A case study enrolled case individuals with glioma (a type of brain cancer) and similar control individuals without glioma. Participants were asked whether they had been regular cell phone users for at least a year. We use logistic regression to model glioma status (1 = glioma, 0 = no glioma) as a function of regular cell phone use (1 = yes, 0 = no). Software gives the following information:
For log odds of glioma/no glioma
$$\begin{array} { l c c }
\text { Predictor } & \text { Coef } & \text { SE Coef } \
\text { Constant } & - 0.08074 & 0.04987 \
\text { Regular use } & - 0.02565 & 0.07275
\end{array}$$ We want to know if the slope of the logistic regression model is significant. What is the value of the test statistic?

Question

A case study enrolled case individuals with glioma (a type of brain cancer) and similar control individuals without glioma. Participants were asked whether they had been regular cell phone users for at least a year. We use logistic regression to model glioma status (1 = glioma, 0 = no glioma) as a function of regular cell phone use (1 = yes, 0 = no). Software gives the following information:
For log odds of glioma/no glioma
$$\begin{array} { l c c } 
\text { Predictor } & \text { Coef } & \text { SE Coef } \
\text { Constant } & - 0.08074 & 0.04987 \
\text { Regular use } & - 0.02565 & 0.07275
\end{array}$$ We want to know if the slope of the logistic regression model is significant. What is the value of the test statistic?

Quizplus · Accepted Answer

The test statistic for the slope in logistic regression is calculated as the coefficient divided by its standard error, which is $-0.02565 / 0.07275 = -0.35$. Since logistic regression typically uses the z-test for significance testing of coefficients, the correct answer is $z = -0.35$.

A Case Study Enrolled Case Individuals with Glioma (A Type